[MightyMouse]

Has anyone ever scaled out of a losing position in just the opposite way you scale into a position?

In a day trade, it would seem to make sense to have as many contracts on as possible when you are going to be right and as few on when you are going to be wrong. Going “all in” and then “all out” at a stop loss tends to cause big swings in your P/L when there’s high volatility. “Scaling in” minimizes your loss somewhat, but minimizes your gain too if you can’t get the full position on.

Scaling out of a loser would seem to have the attractive effect of your right positions having the max contracts on and your losing positions having fewer contracts on than your max. Mathematically, over the same range of prices, in the case of a full loss, the “loser scale out” loss would be somewhere in between the “scale in” and “all in” trades. In the case where you are correct right away, your gain would equal the “all in” gain and be greater than the “scale in” gain. The tricky trades would seem to be the scratch type trades.

Anyway, I have never read of anyone doing this and was wondering if anyone has experience with it or have heard of it in any shape or form.

[maildigger]

When prices are between my entry & my stop I still believe things will work out for the trade.

[Head2k]

I guess it depends on logic of your entries and levels of confidence (if you are able to define them). Maybe the reason why this approach is not popular is that people tend to keep their stops tight and there is no room for scaling out? Maybe they rather exit their full position and then re-enter full again? But if somebody likes wider stops then this approach could be applied. For example, if you buy 3 contracts at an upside breakout from a narrow range, you could place one stop below range resistance, one below midpoint and one below support. This would have some logic. As price would drop after your entry you would decrease size with decreasing confidence in the upward continuation.
Just ideas of a beginner. But to find out whether this approach is beneficial for your type of entries there is only one way: to test it.

[MightyMouse]

Your confidence level of your position and what the market will do with your position are two completely separate things. A market that is turning your position into a loser will continue to turn your position into a loser whether you are confident about the trade working out or not.

Somehow, when you go long for example, you are buying at a level where there is support because in situations similar to this one in the past, support has held up. However, the fact is that, though this situation may look similar to one's you have experienced in the past, it is not going to be exactly like the one's in the past; otherwise, we would all have a “set it and forget it” algorithm and have nothing to talk about. But, you’re really only estimating support and you can be anywhere from 100% right to 0% right about whether your entry was at support. Unfortunately, you can be 100% right about your entry being support, but if you are trading in the wrong timeframe, you could still have a loser.

Sure, when you get filled, if the position moves a little bit against you, there’s no reason to panic. But, say, for example, that you are day trading and your signals are generated from a 5 minute timeframe. If the average range of a 5 minute bar is 20 ticks and your stop is 20 ticks and the position has moved against you 10 ticks, you have to agree that you are much more likely now to get stopped out of your position by white noise movements of the market. Maintaining the same level of confidence, here, may cloud one’s ability to be objective. If you were still truly confident about the position, given the new market information (the market has brought you 10 ticks closer to being stopped out with a 20 tick loss), wouldn’t you want to move your stop back another 10 ticks to compensate for volatility, giving you a potential loss of 30 ticks if you are still as confident? That is never a good idea.

The point of this thread is to see if we can figure out a way to better maximize profits by minimizing costs. One of the greatest costs in our business is our losses.

[BlowFish]

Originally Posted by MightyMouse »

Has anyone ever scaled out of a losing position in just the opposite way you scale into a position?

In a day trade, it would seem to make sense to have as many contracts on as possible when you are going to be right and as few on when you are going to be wrong. Going “all in” and then “all out” at a stop loss tends to cause big swings in your P/L when there’s high volatility. “Scaling in” minimizes your loss somewhat, but minimizes your gain too if you can’t get the full position on.

Scaling out of a loser would seem to have the attractive effect of your right positions having the max contracts on and your losing positions having fewer contracts on than your max. Mathematically, over the same range of prices, in the case of a full loss, the “loser scale out” loss would be somewhere in between the “scale in” and “all in” trades. In the case where you are correct right away, your gain would equal the “all in” gain and be greater than the “scale in” gain. The tricky trades would seem to be the scratch type trades.

Anyway, I have never read of anyone doing this and was wondering if anyone has experience with it or have heard of it in any shape or form.

You might want to take a look at 'Phantom of the Pits' if this interests you. It's a free ebook. There is a short version and a long version.

[MightyMouse]

Originally Posted by BlowFish »

You might want to take a look at 'Phantom of the Pits' if this interests you. It's a free ebook. There is a short version and a long version.

I read that a few months ago. It was a good and his rules make a lot of sense, but he is a long term positional trader and gives you great advice if you intend to build a position to take advantage of a longer term trend. But, day trading is a very different animal. IMO you really can't apply both of his rules to day trading, unless I am missing something.

However, his thought about “not letting the market prove your position wrong” is something to the effect of what I am after. In that, your losing positions when your stops are hit are not as big as your winning positions.


Thanks for the reply.

[steve46]

In markets that can be characterized as reverting to a mean, scaling in and out could work. The problem traders must overcome is that when you are wrong and price continues to trend against your position, your losses can quickly become significant. Since retail traders often have insufficient capital to begin with, unless you have a strong mathematical advantage, one or two strong moves against you will probably take you out of the game.

[BlowFish]

Originally Posted by MightyMouse »

I read that a few months ago. It was a good and his rules make a lot of sense, but he is a long term positional trader and gives you great advice if you intend to build a position to take advantage of a longer term trend. But, day trading is a very different animal. IMO you really can't apply both of his rules to day trading, unless I am missing something.

However, his thought about “not letting the market prove your position wrong” is something to the effect of what I am after. In that, your losing positions when your stops are hit are not as big as your winning positions.


Thanks for the reply.

Yes, perhaps not much help to a scalper. Reminiscences is another good read. You get lines like (paraphrased) 'I let the market have 200 shares to see how it took it'. I love that book. One lesson it provides is the perils of over leverage and using un-realised profits to 'upside down' pyramid.